dl 9-29-15.
1. Preliminaries.
    -We used Stata 13.1 (Windows) for the regressions and sensitivity analyses. 
    -We used R 3.2.2 (Windows) to evaluate balance.
    
    -The following Stata .ado and R packages are required for replication:
        -arsens.ado (included in "ado" subfolder in the replication materials)
        -arsimsens.ado (included in "ado" subfolder)
            -arsens and arsimsens should be placed in a  subfolder where Stata will find them--i.e., where .ado files starting with "a" are located.  Type "adopath" at Stata command prompt if unsure. C:\ado\personal is a typical location.  
        -The .ado files in CLARIFY (including estsimp, setx, simqi, and sumqi--included as a .zip file in "ado" subfolder)
        -The .ado files in spost13.  If not installed, can be installed by typing in the Command window:
           " net from http://www.indiana.edu/~jslsoc/stata/ "
         and then   
           " net install spost13_ado "
        -The R packages foreign and RItools (and its dependencies).   
    
    -IMPORTANT: BEFORE RUNNING ANY .DO FILES, set the local `path' to the path (folder location) where this folder of replication files is located on the computer you are using for replication. This local can be set at the very top of each .do file. For example, if you don't modify the folder's name, this might be something like: < local path="C:\documents\bl_psrm_replication_files" >. Other locals can be left alone.  ALSO: BEFORE RUNNING THE R SCRIPT for balance, modify the path(location) in the setwd() command in the same way.  (In both cases, do not include slash after last folder.)
    -Keep all of the replication files (except the .ado files) in the same folder.

2. Steps to replicate once required packages are installed and path & setwd is modified (see above).
    A. SENSITIVITY ANALYSES.  
        -Tables 1-3. Run "ssa_A5F.do". (Note that this--and the other .do files--will modify (i.e., reconstruct) some of the datasets (.dta) files already in the replication folder. Also note that this .do file may take a fair amount of time before it finishes running.) We log this in "sares_A5F_9-28-15.smcl".
        -Supplemental Tables 7-12. Run "ssa_A56D.do". The associated log is "sares_A56D_9-28-15.smcl".
        -Supplemental Tables 13-15. Run "ssa_A6F.do". (Note that this .do file may take a fair amount of time before it finishes running.) The associated log is "sares_A6F_9-28-15.smcl".
        -To quickly and compactly list these results as tables without (or after) running the sensitivity analyses, run "quickly_list_tables.do".  
    
    B. REGRESSIONS, PREDICTED PROBABILITY GRAPH, and SECOND DIFFERENCE CALCULATION.
        -The regression in Table 4, the predicted probability graph in Figure 2 (and the associated regressions in Supplemental Table 16), and the second difference reported in fn 26 can be replicated by running "combined_regressions.do" This replication is logged in "regreps_log_9-28-15.smcl".
    
    C. BALANCE. 
        -To replicate the results presented in Supplemental Tables 1-6, run "R_RI_Balance_Check.R".  The associated log is "all_balance.log". The short variable names are:      
            usat=US Appellant 
            usae=US Appellee
            sgat=SG Appellant  
            sgae=SG Appellee  
            dcat=DC Elite Appellant 
            dcae=DC Elite Appellee
            lpat=Law Prof. Appellant 
            lpae==Law Prof. Appellee
            clat=Clerk Appellant 
            clae=Clerk Appellee
            elat=Elite Law School Appellant
            elae=Elite Law School Appellee
            li=Liberal Lower Court Decision  
            ex=Relative Experience
            co=Case Complexity 
            me=Court Median Ideology
        -The order in which the table (columns) appear in the log:
            1. Supplemental Table 1, Column 1.
            2. ST 1, Column 2
            3. ST 2, Column 1
            4. ST 2, Column 2
            5. ST 1, Column 1 (again)
            6. ST 1, Column 3
            7. ST 2, Column 1 (again)
            8. ST 2, Column 3
            
            9. Supplemental Table 3, Column 1.
            10. ST 3, Column 2
            11. ST 4, Column 1
            12. ST 4, Column 2
            13. ST 3, Column 1 (again)
            14. ST 3, Column 3
            15. ST 4, Column 1 (again)
            16. ST 4, Column 3
            
            17. Supplemental Table 5, Column 1.
            18. ST 5, Column 2
            19. ST 6, Column 1
            20. ST 6, Column 2
            21. ST 5, Column 1 (again)
            22. ST 5, Column 3
            23. ST 6, Column 1 (again)
            24. ST 6, Column 3

3. List and description of files.
    A5Dsares.dta  SSA results, ptr-better, Specification 1.
    A5Fsares.dta  SSA results, ptr-better, Specification 2.
    A6Dsares.dta  SSA results, resp-better, Specification 1.
    A6Fsares.dta  SSA results, resp-better, Specification 2.
    ado  Folder with .ado files.
    all_balance.log  Log of Balance Checks.
    combined_regressions.do  Script to replicate regressions, predicted probability graphs, and second difference
    fig2.gph  Figure 2 image, Stata format.
    fig2.png  Figure 2 image, PNG format.
    main.dta  Dataset used for regressions and other Section 5 analyses 
    MRA5Dct2sd0tmn.dta  Dataset with matched observations (ptr-better, positive-diff, specification 1) 
    MRA5Dct2sd100tmn.dta  Ds w/ matched observations (ptr-better, large-diff, specification 1)
    MRA5Dct2sd50tmn.dta  Ds w/ matched observations (ptr-better, medium-diff, specification 1)
    MRA5Fct2sd0tmn.dta  Ds w/ matched observations (ptr-better, positive-diff, specification 2)
    MRA5Fct2sd100tmn.dta  Ds w/ matched observations (ptr-better, large-diff, specification 2)
    MRA5Fct2sd50tmn.dta  Ds w/ matched observations (ptr-better, medium-diff, specification 2)
    MRA6Dct2sd0tmn.dta Ds w/ matched observations (resp-better, positive-diff, specification 1)
    MRA6Dct2sd100tmn.dta  Ds w/ matched observations (resp-better, large-diff, specification 1)
    MRA6Dct2sd50tmn.dta  Ds w/ matched observations (resp-better, medium-diff, specification 1)
    MRA6Fct2sd0tmn.dta  Ds w/ matched observations (resp-better, positive-diff, specification 2)
    MRA6Fct2sd100tmn.dta  Ds w/ matched observations (resp-better, large-diff, specification 2)
    MRA6Fct2sd50tmn.dta   Ds w/ matched observations (resp-better, medium-diff, specification 2)
    quickly_list_tables.do  Script to quickly list sensitivity analysis results
    README.txt  This file.
    regreps_log_9-28-15.smcl  Log of regressions and other Section 5 analyses (incl Supp. Tab 16)
    R_RI_Balance_Check.R  Script for Balance Checks
    sares_A56D_9-28-15.smcl  Log for Sensitivity Analyses (specification 1)
    sares_A5F_9-28-15.smcl  Log for Sensitivity Analyses (specification 2, ptr-better)
    sares_A6F_9-28-15.smcl  Log for Sensitivity Analyses (specification 1, resp-beter)
    ssa_A56D.do  Script for Sensitivity Analyses (specification 1)
    ssa_A5F.do  Script for Sensitivity Analyses (specification 2, ptr-better)
    ssa_A6F.do  Script for Sensitivity Analyses (specification 2, resp-better)
    XBCA5Dct2sd0tmn.dta  Dataset with matched observations and covariates (ptr-better, positive-diff, specification 1)
    XBCA5Dct2sd100tmn.dta  Ds w/ matched obs & covariates (ptr-better, large-diff, spec. 1)  
    XBCA5Dct2sd50tmn.dta  Ds w/ matched obs & covariates (ptr-better, medium-diff, spec. 1)
    XBCA5Fct2sd0tmn.dta  Ds w/ matched obs & covariates (ptr-better, pos-diff, spec. 2)
    XBCA5Fct2sd100tmn.dta Ds w/ matched obs & covariates (ptr-better, large-diff, spec. 2)
    XBCA5Fct2sd50tmn.dta Ds w/ matched obs & covariates (ptr-better, medium-diff, spec. 2)
    XBCA6Dct2sd0tmn.dta  Ds w/ matched obs & covariates (resp-better, pos-diff, spec. 1)
    XBCA6Dct2sd100tmn.dta  Ds w/ matched obs & covariates (resp-better, large-diff, spec. 1)
    XBCA6Dct2sd50tmn.dta  Ds w/ matched obs & covariates (resp-better, medium-diff, spec. 1)
    XBCA6Fct2sd0tmn.dta  Ds w/ matched obs & covariates (resp-better, pos-diff, spec. 2)
    XBCA6Fct2sd100tmn.dta  Ds w/ matched obs & covariates (resp-better, large-diff, spec. 2)
    XBCA6Fct2sd50tmn.dta  Ds w/ matched obs & covariates (resp-better, medium-diff, spec. 2)
     